1. Field of the Invention
This invention relates to novel pharmaceutically active compounds which are inhibitors of metalloproteases. Pharmaceutical compositions comprising these compounds as well as methods of their use for treating or controlling disease states or conditions associated with such enzymes are also described.
2. State of the Art
Metalloproteases are involved in a large number of disease states and other conditions in human and other animals. The metalloproteases are a family of enzymes containing zinc at the active site, which facilitates the catalytic hydrolysis of various protein substrates. A subfamily of the metalloprotease family is known as the matrix metalloproteases because these enzymes are capable of degrading the major components of articular cartilage and basement membranes. The matrix metalloproteases include stromelysin, collagenase, matrilysin and gelatinase, among others.
Stromelysin (aka. proteoglycanase, matrix metalloproteinase-3, MMP-3, procollagenase activator, "transin"), collagenase (aka. interstitial collagenase, matrix metalloproinase-1, MMP-1, type II collagenase), and gelatinase (aka. type IV collagenase, matrix metalloproteinase-2, MMP-2, 72 kDa-gelatinase or type V collagenase, matrix metalloproteinase-9, MMP-9, 95 kDa-gelatinase) are metalloendoproteinases secreted by fibroblasts and chondrocytes, and are capable of degrading the major connective tissue components of articular cartilage or basement membranes. Human rheumatoid synovial collagenase is approximately 50% identical to human stromelysin (Whitham et al., Biochem. J., 240:913-916 (1986)). Gelatinase (MR .about.72,000) has been isolated from rheumatoid fibroblasts (Okada et al., Eur. J. Biochem., 194:721-730 (1990)). A higher molecular weight gelatinase (MR .about.95,000; aka. type-V collagenase, matrix metalloproteinase-9, MMP-9) is also secreted by fibroblasts and monocytes and may be involved in cartilage degradation.
Metalloproteases are apparently involved in several arthritis conditions, including osteoarthritis (OA) and rheumatoid arthritis (RA). These diseases are largely due to the loss of articular cartilage. Elevated levels of stromelysin and collagenase have been detected in joints of arthritic humans and animals (Hasty et al., Arthr. Rheum., 33:388-397 (1990); Krane et al., In The Control of Tissue Damage, A. B. Glauert (ed.), Elsevier Sci. Publ., Amsterdam, 1988, Ch. 14, pp. 179-195; Blanckaert et al., Clin. Chim. Acta, 185:73-80 (1989)). Each enzyme is secreted from these cells as an inactive proenzyme which is subsequently activated. There is evidence that stromelysin may be the in vivo activator for collagenase and gelatinase, implying a cascade for degradative enzyme activity (Fo et al., Arch. Biochem. Biophys., 267:211-216 (1988); Murphy et al., Biochem. J., 248:265-268 (1987); Ogata et al., J. Biol. Chem., 267:3581-3584 (1992)). The synthesis of the gelatinase proenzyme is not coordinately regulated with the other two metalloproteinases. The role of gelatinase in the tissue destruction of articular cartilage appears different from the other two enzymes.
Stromelysin and collagenase are also implicated in the articular cartilage damage associated with septic arthritis. Bacterial infections of the joints can elicit an inflammatory response that may then be perpetuated beyond what is needed for removal of the infective agent, resulting in permanent damage to structural components. Bacterial agents have been used in animal models to elicit an arthritic response with the appearance of proteolytic activities (Case et al., J. Clin Invest., 84:1731-1740 (1989); Williams et al., Arthr. Rheum., 33:533-541 (1990)).
Secreted proteinases such as stromelysin, collagenase, and gelatinase play an important role in processes involved in the movement of cells during metastatic tumor invasion. Indeed, there is also evidence that the matrix metalloproteinases are overexpressed in certain metastatic tumor cell lines. In this context, the enzyme functions to penetrate underlying basement membranes and allow the tumor cell to escape from the site of primary tumor formation and enter circulation. After adhering to blood vessel walls, the tumor cells use these same metalloendoproteinases to pierce underlying basement membranes and penetrate other tissues, thereby leading to tumor metastasis.
Periodontal diseases such as gingivitis are also characterized by metalloprotease expression. Both collagenase and stromelysin activities have been isolated from fibroblasts isolated from inflamed gingiva (Uitto et al., J. Periodontal Res., 16:417-424 (1981)). Enzyme levels have been correlated to the severity of gum disease (Overall et al., J. Periodontal Res., 22:81-88 (1987)).
Stromelysin has been implicated in the degradation of structural components of the glomerular basement membrane (GBM) of the kidney, the major function of which is to restrict passage of plasma proteins into the urine (Baricos et al., Biochem. J., 254:609-612 (1988)). Proteinuria, a result of glomerular disease, is excess protein in the urine caused by increased permeability of the GBM to plasma proteins. The underlying causes of this increased GBM permeability are unknown, but proteinases including stromelysin may play an important role in glomerular diseases.
Metalloproteases may also be involved in the rupturing of atherosclerotic plaques leading to coronary thrombosis. The tearing or rupturing of atherosclerotic plaques is the most common event initiating coronary thrombosis. Destabilization and degradation of the connection tissue matrix surrounding these plaques by proteolytic enzymes or cytokines released by infiltrating inflammatory cells has been proposed as a cause of plaque fissuring. Such tearing of these plaques can cause an acute thrombolytic event as blood rapidly flows out of the blood vessel. High levels of stromelysin messenger RNA have been found to be localized to individual cells in atherosclerotic plaques removed from heart transplant patients at the time of surgery (Henney et al., Proc. Natl. Acad. Sci. USA, 88:8154-8158 (1991)).
Degenerative aortic disease associated with thinning of the medial aortic wall is another condition in which matrix metalloproteases may play a role. Aneurysms are often associated with atherosclerosis in this tissue. Increased levels of the matrix metalloproteinases have been identified in patients with aortic aneurysms and aortic stenosis (Vine et al., Clin. Sci., 81:233-239 (1991)).
Expression of metalloproteinases, including stromelysin and collagenase, is observed in unfertilized eggs and zygotes and at further cleavage stages and increased at the blastocyst stage of fetal development and with endoderm differentiation (Brenner et al., Genes & Develop., 3:848-859 (1989)). By analogy to tumor invasion, a blastocyst may express metalloproteinases in order to penetrate the extracellular matrix of the uterine wall during implantation. In addition, evidence exists that collagenase is important in ovulation processes. Collagenase apparently facilitates penetration of a covering of collagen over the apical region of the follicle, allowing the ovum to escape. There may also be a role for strome-lysin activity during ovulation (Too et al., Endocrin. 115:1043-1050 (1984)).
Proteolytic processes have also been observed in the ulceration of the cornea following alkali burns (Brown et al., Arch. Ophthalmol., 81:370-373 (1969)). Collagenolytic and stromelysin activity have also been observed in dystrophobic epidermolysis bullosa (Kronberger et al., J. Invest. Dermatol., 79:208-211 (1982); Sawamura et al., Biochem. Biophys. Res. Commun., 174:1003-1008 (1991)).
In addition to degrading structural components of the extracellular matrix, stromelysin can degrade other in vivo substrates, including the .alpha..sub.1 -proteinase inhibitor, and may therefore influence the activities of other proteinases such as elastase (Winyard et al., FEBS Lett., 279(1):91-94 (1991)).
Because metalloproteases play a role in so many diseases and other conditions, inhibitors of these enzymes have been studied as possible therapeutic agents. In vitro experiments measuring the effect of matrix metalloendoproteinase inhibitors on proteoglycan release from rabbit cartilage explants suggest that stromelysin inhibition may be effective in preventing articular cartilage degradation (Caputo et al., J. Orthopedic Res., 6:103-108 (1988)). Evidence also suggests that inhibitors of stromelysin, collagenase, and gelatinase will be useful to control tumor metastasis (Matrisian et al., Proc. Natl. Acad. Sci. USA, 83:9413-9417 (1986); Wilhelm et al., Proc. Natl. Acad. Sci. USA, 84:6725-6729 (1987); Liotta, et al., Lab. Invest., 49:636-649 (1983); Reich et al., "Inhibitors of collagenase IV and cell adhesion reduce the invasive activity of malignant tumor cells", in Metastasis: Ciba Foundation Symposium; Wiley, Chichester, 1988, pp. 193-210). An inhibitor of collagenase has been shown to be effective in preventing ovulation (Woessner et al., Steroids, 54:491-499 (1989)). Mercapto-containing peptides inhibit the collagenase isolated from alkali-burned rabbit cornea (Burns et al., Invest. Ophthalmol., 30:1569-1575 (1989)).
Thiol carboxylic acid derivatives that inhibit collagenase are disclosed in U.S. Pat. Nos. 5,109,000; 4,595,700; 4,371,466. Additional collagenase inhibitor compounds are disclosed in European Patent Application Publication Nos. 0 423 943; 0 273 689; 0 322 184; and 0 185 380, and in International Patent Application Publication Nos. WO 88/06890 and WO 94/07481.
Collagenase inhibitors have also been designed around the cleavage site of the a-chain sequence of Type II collagen (Johnson et al., J. Enzym. Inhib., 2:1-22 (1987)). One such inhibitor, N-[3-(benzyloxy-carbonyl)amino-1-carboxy-n-propyl]-L-leucyl-O-methyl-L-tyr osine, N-methylamide, prepared at G. D. Searle, Inc., is a potent inhibitor of human rheumatoid synovial collagenase (IC.sub.50 =0-8 .mu.M). This compound also inhibits rabbit bone proteoglycanase (IC.sub.50 =0.5 .mu.M) (Delaisse et al., Biochem. Biophys. Res. Commun., 133:483-90 (1985)).
However, significant obstacles continue to stand in the way of clinical exploitation of metalloprotease inhibitors. First, there is very little to guide one in developing a specific inhibitor for each enzyme. In preliminary studies of rabbit proteoglycanase with substrates and inhibitors, little was found to indicate the enzyme's requirements for hydrolysis or inhibition beyond a preference for hydrophobic residues at the P.sub.1 ' position (Shaw et al., Adv. Inflam. Res., 12: 67-79 (1988)). More extensive studies with a series of substrates revealed that stromelysin will tolerate nearly every amino acid residue around the scissile bond (Fields et al., unpublished results presented at the Matrix Metalloproteinase Conference, September 1989, Sandestin, Fla.).
Toxicity is a second obstacle to therapeutic use of previously known metalloprotease inhibitors. For example, certain hydroxamic acids have been suggested as collagenase inhibitors as in U.S. Pat. No. 4,599,361 and European Patent Application Publication No. 0 236 872. U.S. Pat. Nos. 5,304,604, 5,240,958 and 5,310,763 also disclose hydroxamic acid derivatives which act as inhibitors of metalloproteases involved in tissue degradation, such as collagenase, stromelysin (proteoglycanase), gelatinase and collagenase (IV).
Although these hydroxamic acid compounds are effective inhibitors of matrix metalloproteases, the hydroxamic acid moiety is potentially toxic. See, for example, Musser et al., J. Med. Chem., 29:1429-1435 (1986); Baililien et al., Am. J. Vet. Res., 74:2604-2611 (1986); Rodman et al., R.L. Clin. Pharmacol. Ther., 42:346-350 (1987); Williams et al., N. Engl. J. Med., 311:760-764 (1984); Yoshioka et al., Mutat. Res., 170:93-102 (1986); Gillissen et al., Carcinogenesis, 15:39-45 (1994); Gillissen et al., Carcinogenesis, 13:1699-1703 (1992); Fishbein et al., Science, 142:1069-1070 (1963);and Borenfreund et al, J. Nat. Cancer Inst, 32:667-679 (1964). As a result, there are few, if any, hydroxamic acid based drugs in use.
The wide spectrum of clinical indications for matrix metalloprotease inhibitors establishes a clear need for matrix metalloprotease inhibitors that have satisfactory inhibition activity. It is not a simple matter, however, to predict what variations in known compounds would retain or even increase activity. The present invention fulfills this need for novel effective metalloprotease inhibitors.